Chapter 4 - Springer Link

Chapter 4 Manipulating Filamentous Fungus Chemical Phenotypes by Growth on Nutritional Arrays José R. Tormo, Francisco J. Asensio, and Gerald F. Bills Abstract Methods for manipulating and fermenting microorganisms in multi-well plates offer unlimited possibilities for high-throughput parallel experimentation. Furthermore, bar-coded data tracking and downstream processing with modern liquid handling equipment reduce handling errors and are able to format microbial products for autosampler-equipped analytical instruments, e.g., HPLCs, mass spectrometers, and plate readers. An integrated system for high-throughput culturing of filamentous fungi replicating strains across many fermentation parameters, called nutritional arrays, was developed. It takes advantage of this equipment while addressing the age-old dilemma of how to manipulate fungal phenotypes to express a more complete spectrum of their secondary metabolites. Growth of any given strain in a well-designed nutritional array increases the chances of detecting a biologically active metabolite while reducing the manpower and materials needed for preparing individual fermentations and extracts. Fungi fermented in nutritional arrays are directly processed in a semi-automated fashion and the extracts prepared for bioassays and analytical chemistry. The necessary equipment, custom tools, and protocols to grow fungi in nutritional arrays are described along with examples of bioactive secondary metabolites discovered using this system. Key words: Automation, Fungi, OSMAC, Secondary Microextraction, LIMS, Data mining, Drug discovery

metabolites,

Microfermentation,

1. Introduction For decades, scientists analyzing data from high-throughput screening of many fungal and bacterial strains for bioactive secondary metabolites (SMs) have been aware of the unpredictability of fermentation-dependent biosynthesis of SMs. Maximizing the qualitative and quantitative metabolic output of a prospective SM-producing strain has long been a major objective of industrial screening programs searching for new applications of bioactive Nancy P. Keller and Geoffrey Turner (eds.), Fungal Secondary Metabolism: Methods and Protocols, Methods in Molecular Biology, vol. 944, DOI 10.1007/978-1-62703-122-6_4, © Springer Science+Business Media, LLC 2012

59

60

J.R. Tormo et al.

small molecules in medicine and agriculture. However, only during the last decade have the genetics underlying the secondary metabolome and the evolutionary processes driving natural product synthesis been substantially clarified. It is now clear that the number of presumptive SM gene clusters in every filamentous fungal genome examined so far exceeds the number of detectable metabolites from a given species (1–3) indicating the existence of numerous unexpressed pathways possibly or likely encoding the biosynthesis of both known and heretofore undescribed (novel) secondary metabolites. Referred to as “cryptic” or “silent” gene clusters, unexpressed gene clusters might be better called “orphan” gene clusters, because their “silence” is probably an artifact caused by our inability to simulate the environmental conditions or chemical signals required for their expression (4). The value of empirical strategies for manipulating nutritional and environmental parameters to stimulate the metabolic diversity of a microorganism has long been recognized. In industrial microbial screening laboratories, e.g., at Merck Research Laboratories, these strategies were called “media screens.” Media screens, empirical comparisons of an organism’s metabolic response to a large number of medium formulations, over a range of temperatures and aerations, were often the first-line approach for improving product formation and associated bioactivity from a newly discovered fungus or actinomycete. The impact of media screens was also evident during the initial fermentation of large number of new strains and testing extracts across multiple biological assays (5). Review of screening data often recognized that interesting active metabolites were formed or produced in larger quantities in only one of several media. Countless hours were spent debating the hypothetical number of media needed for getting a new metabolite from a newly isolated strain, resulting in the general acknowledgement that several growth conditions were always better than one. However, when faced with the task of manually preparing, labeling, inoculating, and extracting many flasks, the practical reality set in that the number of possible conditions for growing each organism was limited. These empirical methods were later formalized in a paradigm designated, one strain, many compounds (OSMAC) (6–8). OSMAC articulated a strategy for modulating the metabolism of poorly characterized microorganisms, often with unknown nutrition, genetics, and taxonomy. Statistical models of the effects of increased number of media on the detection of antifungal signals from extracts have confirmed that the OSMAC approach indeed increases the probability of detecting antimicrobial products from a given strain (7, 9). The major difficulty with OSMAC is that open-ended testing of fermentation media, temperatures, agitation rates, and other parameters and subsequent product analysis can be laborious and expensive. However interpretation of annotated genomic data can now inform what types and numbers of

4

Manipulating Filamentous Fungus Chemical Phenotypes…

61

metabolites can be expected, and miniaturized fermentations and robotics can facilitate the work of finding optimal conditions for their expression. By the early 2000s, parallel experimentation with microfermentations of bacteria and yeasts in multi-well plates (MWPs) had become widely accepted (10, 11). Effective MWP fermentation systems, pin tool inoculators, and colony picking robots became available for use with heterogeneous strain collections of aerobic bacteria and actinomycetes, and clone libraries in Escherichia coli and yeasts. However, when our laboratory considered implementing a new high-throughput strategy to prescreen filamentous fungi under many nutrient regimes and amplifying their ability to produce antifungal or antibacterial metabolites (Fig. 5), we found few precedents that suggested that filamentous fungi could be effectively handled in MWPs, or much less that they would produce their SMs in microplate wells. When first challenged with the idea, several complications were evident. Most fungi in agar culture lack spores and grow only as multicellular filaments and therefore could not be transferred as homogeneous microliter volumes. Mixed populations of fungi vary widely in their rates and patterns of hyphal growth. Aggressively growing species may produce enough biomass to overgrow and escape from microwells. When transferring inoculum or pipetting extracted mycelia, hyphal masses clog and interfere with pipette tips. Furthermore, orbital shaking to aerate microwell cultures would be futile because mycelia quickly grow into solid masses. Therefore, we improvised and adapted existing protocols for MWP fermentations of bacteria and actinomycetes to overcome and accommodate these peculiar behaviors of filamentous fungi, particularly those fungi that only grow as hyphae in culture (9, 10). We adapted a commercially available spring-loaded replicator, and sandwich covers for deep well MWPs for inoculation and growth of filamentous fungi in 96-well and 24-well formats (www. enzyscreen.com; www.kuhner.com). A technique was developed for efficiently generating inoculum from fungal hyphae and transferring inocula among microwells as hyphal suspensions (Figs. 1, 2, and 3) (9). A simple screening system was established using heterogeneous groups of 80 fungi grown in 96-well format replicating hyphal suspensions across multiple media. Fungi were incubated statically to simulate solid-state fermentations (Fig. 4). Fungi could be efficiently extracted from MWPs, and metabolite production was adequate for detection of biological activity (Fig. 5). Liquid handling robots had many different incompatibilities with direct extraction in MWPs and presented significant obstacles achieving the desired throughput. These obstacles were overcome by designing a new system that allowed solvent dispensing and direct pipetting from MWPs and the use of disposable tips for extract handling. A custom tool (Figs. 8 and 9) was built to compress

62

J.R. Tormo et al.

Fig. 1. Preparation of fungal nutritional arrays. Fungal hyphal suspensions prepared in tubes with cover glasses.

Fig. 2. Preparation of fungal nutritional arrays. Rack of hyphal suspensions organized in the same pattern as the master plate.

fungal cells to the bottom of deep well microplates enabling clean pipetting of extracts (10). Growth in 96-well plates could yield 500–1,000 ml of mycelial extract and supernatant, enough to carry out several assays, repeat them if necessary, and evaluate the extract via HPLC-MS. Active strains discovered by high-throughput screening of microplate fermentations usually can be scaled up in shaken flask or static flask cultures in order to isolate and characterize the active components.

4


63

Fig. 3. Preparation of fungal nutritional arrays. Inoculating a medium plate with 80 fungi.

Fig. 4. Preparation of fungal nutritional arrays. Incubation of nutritional arrays in an incubator. Plates on upper shelf are closed with carpenter clamps. Plates on bottom shelf are closed with custom-built supports.

Because of the potential to generate masses of screening data, a suitable data-mining environment is recommended for getting the most out of the nutritional array methodology, especially for making comparisons in assay responses among fungus × medium combinations (e.g., Fig. 5). Automation of laboratory procedures effectively increases the chances of successfully implementing innovative solutions. Computer-assisted follow-up of integrated

64

J.R. Tormo et al.

Fig. 5. Prescreening of 800 extracts from a fungal nutritional array with a zone of inhibition assay against Staphylococcus aureus. Eighty fungi were replicated and grown in ten different media in the central 80 wells of 96-well plates. Extracts were applied to preformed wells in Omnitray plates (Nunc) seeded with S. aureus. Positions second and seventh from the first columns per plate are positive controls, kanamycin and vancomycin, respectively. Images were captured by a custom image analyzer, and zones of inhibition were outlined and measured by the analysis program. Media designations are in white letters.

processes, given the number of strains, media, fermentations, and extracts that grow factorially with nutritional array methodology, substantially improves data tracking and interpretation. An internal computer network environment for managing users and instruments with access control and method monitoring will help to get the most of these programs. Connecting liquid handling robots

4


65

and assay readers enables both storage of valuable data and direct access to processed data. These operations can be managed by determining the application needed for basic operational use of each piece of laboratory equipment, plus the design and development of specific procedures needed for each of the nutritional array protocol steps. Correct configuration of each instrument and installation of integrated data transfer solutions enable a productive integration of the different steps and the automated storage of the data for its processing and decision making. Storage of data regarding strains, fermentation media, solvent extracts, well positions, plates, volumes, biological results, chemical results, and standard operating procedures (SOPs) that mirror each step in large-scale research programs is commonly approached by the use of databases. Our data tracking employed Oracle Database because of its robustness, simplicity, and ability to create distributive networks that facilitate integration of each component of the system and sharing of information with other research environments (users, laboratories, departments, companies, and institutes). The connectivity offered by Oracle DB allows the use of a large series of client/Web-server applications for processing, visualization, or sharing of data from the database that, in coordination with a file server, will allow effective distribution of data between the different pieces of equipment, laboratory technicians, and scientists. Security of data is also recommended with specific SOPs that ensure duplication and integrity of each data point generated from every fungus × medium combination. Below we describe the necessary equipment, custom tools, protocols to grow fungi in nutritional arrays, and data management for recording and analyzing results. First-time users should understand that our system evolved in a high-throughput pharmaceutical screening environment; however most of the liquid handling steps and software solutions can be accomplished with simpler tools, e.g., multichannel pipettes, peristaltic pumps coupled to manifolds, and common small-business data-mining applications. We encourage anyone interested in maximizing the full potential of fungi to express their range of chemical phenotypes to experiment with nutritional arrays and design their own microfermentation experiments.

2. Materials 2.1. Labware Infrastructure and Equipment

1. Autoclave for sterilizing tools, media, and plates. 2. Climatic chamber or controlled temperature room for maintaining stable fermentation conditions. Mesophilic fungi are normally grown between 20 and 30°C. If possible, relative

66

J.R. Tormo et al.

humidity should be maintained at 60% or higher, especially in arid climates. 3. Freezers (−80°C or liquid N2) for cryopreservation of fungal strains. 4. Stereo- and bright field microscopes for checking purity and identity of strains. 5. System Duetz (www.enzyscreen.com; www.kuhner.com), a set of tools developed for miniaturized parallel preservation and aerated cultivation of bacterial strains in microplates with 96-, 48-, or 24-well formats. 6. Rotary shaker (e.g., New Brunswick Scientific, USA; Kühner AG, Switzerland; or other manufacturers) with clamps for aerated culturing of inoculum and shaking platforms (5 cm displacement) with inclinable (approximately 75° angle) steel racks (25 mm diameter, 16 tubes per rack) for incubating and agitating seed inoculum tubes. 7. Level P2 biological safety cabinet for culture and media manipulations. 8. Ventilated chemical safety cabinet for the handling of volatile organic solvents, extraction of fermentations, and solvent evaporation. 9. Regulatory approved disposal system for biohazards and used chemicals. 10. Small liquid handling tools, e.g., multichannel pipettes, Thermo Multidrop Combi. 11. Liquid handling station (e.g., Tecan Genesis, Biomek FX) for the distribution of extracts in 96-well assay plates. 12. Barcode printers and readers for plate tracking. 13. Low-speed vacuum centrifuges, e.g., Genevac HT-24 (Genevac Inc.) computer-controlled vacuum centrifuge, or Savant Speedvac Plus (SC2010A) for solvent evaporation in MWPs. 14. A custom aluminum plunger that filters and squeezes the mycelia from the walls into the bottom of the plate. Alternatively, a high-speed centrifuge for MWPs can be used to partially separate fungal cells from extracts. 15. Database or spreadsheet programs for barcode labeling and tracking of strains, strain × medium combinations, plates, plate coordinates, extracts, and assay results. 2.2. Information Technologies Infrastructure and Equipment

1. Data mining hardware: Different options for servers are available, with a range of quality-to-cost ratios. We selected a specific solution (HP, www.hp.com) because of our long experience and confidence in its reliability, e.g., 4× ProLiant DL360 G5 with 2 Xeon E5405 2 GHz processors, 64 Gb of RAM

4


67

memory, 2×146 Gb hard drives for operative systems (OS) and virtualization software, plus two hard drive cabinets with 11 450 Gb discs for a total of 4 Tb scalable up to 21.6 Tb with two additional hard drive cabinets. 2. Data-mining configuration: Modern IT hardware allows the distribution of computer flows virtually among physical servers in a balanced way. Moreover, hardware power can be divided in isolated virtual servers for better performance. HTS databases should be placed, if possible, in one of these virtual divisions to reduce chances of failure and data corruption. 3. Storage of data from databases in hard drives can also follow the same paradigm of dividing physical space (e.g., 2× 450 Gb hard drives) into virtual units with redundancy of data in case of technical failure. The combination of virtual servers and storage of data results in a scalable and highly versatile programming environment that can evolve with the research projects as they evolve. 2.3. Applications and Software Solutions

1. Databases: Two databases, production and development, are commonly needed to allow the implementation of software applications and laboratory procedures while laboratory operations evolve in response to experimental results; the two databases ensure security and functional structure for the data. 2. Database Software: Oracle Database (www.oracle.com) has been our preferred solution. Oracle on Linux ensures robustness, good design, easy learning curve, and administration, and has the largest capabilities of online maintenance and interaction with other external or previous historical databases. 3. Laboratory Information Management System (LIMS): LIMS are central applications essential for tracking and recording experiments and related manipulations (sample preps, equipment quality control, etc.) involved in all steps of a research process (Fig. 6). We created an internal LIMS (Culture Tracking System or CTS) with a nutritional array-specific module for supporting this complex fermentation scheme and other programs related to the fermentation of strains. Other commercial options, e.g., ActivityBase (IDBS, www.idbs.com) or Nautilus (Thermo Scientific, www.thermo.com), oriented, respectively, to chemical or biological laboratories, also include a complete series of modules for tracking, calculations, and statistics. 4. Physical Tracking and Documentation: Electronic databases efficiently maintain information, but data input can be a time-consuming task and becomes an important handicap for the implementation of any high-throughput system. Barcode labeling software (e.g., Bartender, Seagull Scientific, www.seagullscientific.com) is recommended for the correct

68

J.R. Tormo et al. Nutritional - Array Plate Samples Plate Assay Plate

Screening

Cherry Picking LC/MS Secondary Assays

Results DB

Stat Analysis

Chem DB

Selection LIMS DB Scale - up

Purification

Fig. 6. LIMS system for centralizing data, workflow tracking, and registering inocula, fermentations, and extracts for high-throughput generation of nutritional arrays. The configuration enables sample tracking, hit selection and comparison, cherry-picking of actives for confirmation and follow-up, and scale-up selection.

identification (human and robot reading) of inocula, fermentations, extractions, and assay plates created by the LIMS system. 5. Visualization tools and statistics: Data-mining applications are needed to summarize and extract information from the extensive raw data. Statistical packages can find patterns among serendipitous data and can discern and control experimental errors. Visualization of data (e.g., Tibco Spotfire, www.tibco. com) can aid in comparisons between fermentation media for a given population of strains, identification of unique media for the production of new SMs, or simple evaluation of the number and type of strains, fermentations, extracts, or actives from a nutritional-array program. 2.4. Ingredients and Reagents

1. Petri dishes and agar media appropriate for fungi, e.g., malt extract agar, potato-dextrose agar. Omnitray plates (Nunc) for sterility controls. 2. Yeast extract medium (YM): Malt extract 10 g; yeast extract 2 g; bacteriological agar 20 g; distilled H2O 1 l.

4


69

3. Sabouraud’s maltose yeast extract medium: Maltose 40 g; yeast extract 10 g; neopeptone 10 g; agar 3.5 g; distilled H2O 1 l. 4. Cryovials with sterile 20% glycerol and storage boxes for culture preservation. 5. Automatic and disposable pipettes, Transfer tubes (Spectrum Laboratories, Rancho Dominguez, CA). 6. Adhesive labels and marking pens. 7. 25 × 150 mm glass tubes with closures. 8. 22 mm long glass covers. 9. Deep-well polypropylene 96-well (2.4 ml total volume) or 24-well plates. Steel lids, breathable membranes, and silicone closures for plates (see System Duetz). 10. 250-ml Erlenmeyer flasks for medium aliquots. 11. Ingredients for media: Media selection will depend on the kinds of fungi grown and the investigator’s experience and preferences. Because one of the goals of the technique is to take advantage of automated liquid handling, generally media with defined or soluble complex ingredients are used to facilitate medium dispensing. 12. Solvents for extraction: Historically acetone has been used, but other water-miscible solvents could work as well. DMSO for suspension of dried extracts.

3. Methods 3.1. Inoculum Preparation, Replication, and Growth of 80 Fungi in MWPs

1. Select sets of 90–100 strains of newly isolated fungi or fungi from a culture collection and grown in YM agar for 2–3 weeks in 60 mm Petri dishes. Avoid heavily sporulating species with dry airborne spores (e.g., Aspergillus spp.) or species with extremely fast radial growth (e.g., Trichoderma spp.) to minimize the possibility of cross-well contamination (see Note 1). 2. Prepare seed medium in 25 × 150 mm glass tubes. Add two 22-mm-long cover glasses to each tube before adding medium. Agitation on an orbital shaker causes the cover glasses to continually shear hyphae and mycelial pellets resulting in a homogeneous hyphal suspension (Fig. 1). 3. Dispense 8 ml of melted and homogenously stirred Sabouraud’s maltose yeast extract medium with 0.35% agar into each tube. 4. Cap tubes and autoclave. 5. Inoculate the tube with two 5-mm diameter agar discs cut from each culture with a Transfer Tube (Spectrum Laboratories). The Transfer Tube’s plunger is pushed against the bottom of

70

J.R. Tormo et al.

the tube to mash and extrude the agar discs against the glass; this increases the number of mycelial growing points. 6. Agitate tubes at 200 rpm, 5 cm displacement, for 3–6 days, where most strains will exhibit moderate to dense growth. However, tubes that appear to the unaided eye to have little growth, will, upon microscopic inspection, often reveal a dense suspension of fine hyphal fragments. 3.2. Preparation of Master Plate for Replication

1. Select 80 strains from among sets of 100 tubes, with adequate growth for the preparation of the master plates for replication (Figs. 1, 2, and 3). 2. Use a sterilized master plate for holding the hyphal inoculum for replication. Use the Sterilized Transfer Tubes to transfer 1 ml of hyphal suspensions to the wells (see Note 1). 3. Additional aliquots of each inoculum tube in 20% glycerol in 1.8 ml cryovials at −80°C can be frozen after preparation of the master inoculum plate for easy recovery of screening actives. 4. Close the master plate after loading, until replication in fermentation plates is needed.

3.3. Preparation of Fermentation Media in MWPs

1. Prepare the fermentation media in advance, in aliquots of 100 ml in Erlenmeyer flasks. 2. Use 80 ml of the aliquots to fill the 80 central wells of each deep-well growth plate at 1 ml per well (see Note 2). 3. The first and last columns of the plates are left empty, so those positions can be used for evaluation of assay-positive and -negative controls. Media formulations usually consist of soluble components. See Note 3 for solid media recommendations. 4. Close fermentation media plates with the sandwich lids. 5. Autoclave the assembled plates with media and then store them in a refrigerator until use.

3.4. Replication of Master Plates Across the Nutritional Array

1. Sterilize the cryoreplicator pin tool by immersing the pins in 70% ethanol for 1 min and then evaporating the ethanol on a hot plate. 2. Once thoroughly cooled insert the cryoreplicator pins several mm into the hyphal suspensions of the master plate and then raise them. 3. Switch the master plate for a medium-filled growth plate, and lower the inoculum-covered pins to the bottom of the liquid medium. 4. Switch the growth plate again for the master plate, and repeat the inoculation three to five times. Repeated inoculation from the master plate ensures that sufficient fungal cells are transferred to initiate growth.

4


71

5. Repeat the process for each different medium plate. Sterilization of the cryoreplicator during multiple transfers from the same master plate is unnecessary because the cryoreplicator press prevents accidental lateral movements, and the inoculum source for each well remains constant. 6. Change cryoreplicators and sterilize them between consecutive master plate inoculations. 3.5. Incubation

1. Place plates statically at about an 85° angle in a custom-built steel support (Fig. 3). See Note 4 for alternative incubation methods. The support’s thumbscrews seal the silicone layer preventing cross-well contaminations. Inclined incubation increases the surface area of the liquid media and improves aeration. 2. Incubate plates for two- to three-week growth cycles at 22°C.

3.6. Contamination and Viability Check

1. Fill Omnitray plates with YM inoculum agar and inoculate them from the same master plate incubated at 22°C. 2. Fill a second Omnitray plate with Luria broth agar (from Sigma) and inoculate it. 3. Incubate both plates at 28°C to check for bacterial contamination. Replication of master plates onto agar is used in this case to verify that each well contains viable inoculum and that strains are not contaminated.

3.7. Extract Preparation

1. After growth, open plates and inspect them for contamination. 2. Replace the sandwich cover’s 96-hole silicone mat with a solid silicone mat, removing the breathable cotton layer. 3. Dispense a volume of 850 ml of acetone (or other water-miscible solvent) into each fungal culture (see Note 5). 4. Gently dislodge mycelia adhering to the well walls. Crush and mix mycelial masses with the help of the cryoreplicator tool. 5. Reassemble and clamp Sandwich covers with the solid silicone mats to a shaker board and agitate for 1 h. After about 30 min, reverse the shaker orientation to change the direction of agitation. 6. Without the Sandwich covers, reduce the culture-solvent mixture up to its 50% of original volume by vacuum evaporation in a Genevac HT-24 for 1 h (Fig. 7). 7. After solvent removal, approximately 500 ml of aqueous extract remains per well. Plates are centrifuged to settle the mycelia (3,800 rpm, 4036 g, 20 min). 8. Add 3–4 h in a chemical hood after vacuum centrifuge evaporation of organic solvent to achieve a more homogeneous and complete solvent evaporation.

72

J.R. Tormo et al.

Fig. 7. Extracts’ preparation from fungal nutritional arrays. Plates with solvent and mycelia supernatants in GeneVac vacuum evaporator.

9. To facilitate extract pipetting from growth plates, a custom-built aluminum 96-well hollow plunger (Figs. 8 and 9) can be used to compress mycelium to the bottom of wells while the extract and supernatant rise in the plunger’s hollow columns. The hollow center of the plunger columns and open windows at the sides prevent spilling and overflow of the extracts and eliminate mycelial interference during pipetting with a liquid handling robot. 10. Transfer a total of 500 ml of the aqueous-based supernatant from each well of the original 2-ml fermentation MWPs to a 500–800 ml well plate with a Beckman 96 head Biomek FX robot (Fig. 9). This machine was selected for its ability to handle with a conveyor belt the 2-ml heavy plates with the metal plungers from a Cytomat Carrousel to the pipetting platform. Fly-by barcode readers, one for the source plates and a different one for the destination plates, are convenient for crosschecking of source and destination plates (see Note 6 for alternative extraction methods for growth media with insoluble components). 11. Store the samples at −20°C until assay by creating the library of potential sources of new SMs for screening. 12. Before assaying, thaw sample plates (Fig. 10) and briefly shake them for 20 min on a plate mixer. 3.8. Structure of the Data

Data mining of nutritional array experimentation can be a limiting step depending on the dimensions of the experimental data. Small projects can be handled with a calculus spreadsheet and a picture

4


73

Fig. 8. Extracts’ preparation from fungal nutritional arrays. Aluminum plungers used to compress mycelium in deep well plates.

Fig. 9. Extracts’ preparation from fungal nutritional arrays. Pipetting the aqueous phase from inside the metal plunger holes into assay blocks.

of the assay plates (e.g., Fig. 5); larger projects will need an IT database development environment as described in the infrastructure section. Resources in each lab will need to evaluate and determine the quantity of data that will be useful to confirm an experimental hypothesis. In practice, development of nutritional array methodologies would resemble expanding the secondary metabolite production dimension from fermentations of a given group of strains. The SM production will be evaluated by different methodologies such as

74

J.R. Tormo et al.

Fig. 10. Extracts’ preparation from fungal nutritional arrays. Extract sample plate ready for bioassay. Columns 1 and 12 are used for assay controls.

taxonomy, biological assay results or chemical evaluation (Fig. 6); in reality data becomes multidimensional as the number of fermentations becomes wider and a number of analyses become available. No matter whether the infrastructure available is small or large, there are several pieces of information that should be kept independent and open for later interrogation of the data. Experiments will afford more and better conclusions if the raw data is kept organized and available for querying. In general, for each fermentation batch, we suggest, if possible, to track independently: 1. Microorganism code, microorganism substratum (source, county, area), preservation details, microorganism characteristics (e.g., morphology, taxonomy, taxonomy details), inoculum media (composition details), physical inoculum fermentation parameters (e.g., temperature, days, shaken, light, etc.), fermentation media (composition details) and physical parameters (e.g., temperature, length of cycle, agitation rate, light, etc.), fermentation SOP. 2. Extraction code, part of the fermentation extracted (e.g., crude, mycelium, pellet, etc.), kind of extraction (e.g., liquid/liquid partition, solid/liquid extraction, solid-phase extraction, etc.), extraction solvent, final sample characteristics (solvent, concentration, aliquot number, volume), extraction SOP. 3. Plate code, format (e.g., 24 wells, 96 wells, 384 wells, etc.), plate layout (e.g., 80 central positions with extracts leaving the first and last column empty for biological controls), coordinates

4


75

of the extract in the plate (row, column), volume per well of extract, storage history (freeze–thaw cycles), sample management SOP. 4. Biological data per each assay performed: Assay test, replicate, response (e.g., 19 mm halos), response compared to controls (e.g., % vs. controls), characteristics of the response (e.g., turbidity, definition), biological evaluation SOP. 5. Chemical data for each analysis: Chemical test (e.g., LC, MS, LC/MS), characteristics of the test, detailed result in a database format, chemical evaluation SOP. 6. Other evaluation dimensions possible. Depending on the questions that need to be answered from the nutritional array fermentations, the structure of the database will need to be elaborated. Simple questions only will need simple data tracking, but with the situation of fermenting 80 strains with several fermentation media, with several biological or chemical evaluations, the better the data mining tools, the more answers we will be able to extract from the same experimental effort. 3.9. Conclusion

During the 2 years that the nutritional array program was fully operative in our laboratory, more than 163,000 extract samples were prepared from fungi; from those crude extracts 3,900 fungal fermentations were selected for scale-up due to their suitable antibiotic and metabolite profiles (12). The resulting collection of prescreened bioactive samples constituted one of the most valuable collections of natural product extracts generated within an industrial environment (12, 13) and copies of those extracts remain available at Fundación MEDINA (www.medinaandalucia.es). In summary, we find that growth performed in 96 or 24 well plates allow for creative and high-throughput experimentation with fungi that otherwise would be resource-limited and cumbersome with shake-flask systems. Growth among phylogenetically related strains can be quickly compared under identical conditions for their production of bioactive metabolites (12, 14–16). Nutritional arrays enabled the prescreening of thousands of fungal strains for medium conditions producing antifungal and antibacterial metabolites; identification of the optimal culture medium facilitated scale-up for chemical genetic profiling in genome-wide Candida albicans and Staphylococcus fitness tests (12, 13). Profiling of some of these extracts, first generated at 1-ml volumes, led to the identification of new fungal metabolites with novel mechanisms of bacterial and fungal antibiosis (13, 17, 18). A convincing example of the potential of nutritional arrays was the discovery of fellutamides C and D from an unknown Metulocladosporiella sp. (13, 19). The Metulocladosporiella strain’s antifungal phenotype was revealed during a validation experiment comparing the antifungal activity

76

J.R. Tormo et al.

from a set of 320 fungi grown in typical vial fermentations on three media to that from the same fungi grown in a 12-media nutritional array. From the entire set of 15 media, only one extract from a sucrose and yeast extract medium in the nutritional array was antifungal. A larger set of samples was generated by growing multiple wells of the fungus; components of its organic extract were determined to affect genes for the 20S subunit of the C. albicans proteasome, a novel mode of action for an antifungal agent. The biosynthesis of fellutamides C and D was so specific to the medium that it was necessary to mimic the 1-ml microfermentation by growing the fungus statically in a thin layer of the same medium in Fernbach flasks for the production of the compounds (19).

4. Notes 1. Screening applications that need to use heavily sporulating species with dry airborne spores or with very fast growth should consider inoculation patterns that skip alternate rows or wells for avoiding cross-growth contamination. 2. Adding in the medium more agar than indicated reduces mycelial pelleting and aids in adherence of mycelial fragments to the replicator tips. 3. Growth media with insoluble components or precipitates (e.g., CaCO3 or starch) can be used with this system with a magnetic stirrer during dispensing. Seed-based solid media can be prepared by filling an MWP with 200-ml wells with seeds (wheat, rice, millet, cracked corn), leveling the seed depth to the tops of the wells. The open wells of an inverted growth plate can be aligned on top of the seed-containing microtiter plate, and the two can be quickly flipped, so that a measured seed volume drops into each well. Finally add a liquid basal medium (700 ml) to the seeds. See Note 6 for appropriate extraction. 4. An alternative method for plate incubation can be improvised by holding the sandwich covers in place on growth plates with heavy rubber bands or with carpenter’s wood clamps that clamp lids securely to growth plates, and then carefully tilting the plates on the edge of a 4-mm high edge (e.g., two wooden tongue depressors). 5. DMSO can be added during the extraction process to minimize metabolite precipitation during solvent evaporation. A final 20% of DMSO in the sample vehicle (water) is compatible with most of the cell-based assays. 6. Description of extracting protocol covers most of the fungi fermentations. A variant of the protocol was developed in

4


77

response to the presence of excess of starch in the media when processing solid seed-based fermentations (2.5% of our total extracts). In the case of solid fermentations (see Note 3), it is more convenient to add solvents, shake, centrifuge, pipette the supernatant, and evaporate the extracts in an additional barcoded 2-ml deep-well plate before their final transfer to the 500–800 ml extract well plate rather than evaporating solvents in the presence of the mycelia.

Acknowledgements We would like to express our gratitude to all our former coworkers and colleagues who participated in the development, validation, and improvement of the nutritional array method. References 1. Khaldi N, Seifuddin FT, Turner G, Haft D, Nierman WC, Wolfe KH, Fedorova ND (2010) SMURF: genomic mapping of fungal secondary metabolite clusters. Fungal Genet Biol 47:736–741 2. Brakhage AA, Schroeckh V (2010) Fungal secondary metabolites - strategies to activate silent gene clusters. Fungal Genet Biol 48:15–22 3. Hoffmeister D, Keller NP (2007) Natural products of filamentous fungi: enzymes, genes, and their regulation. Nat Prod Rep 24:393–516 4. Gross H (2007) Strategies to unravel the function of orphan biosynthesis pathways: recent examples and future prospects. Appl Microbiol Biotechnol 75:267–277 5. Yarbrough GG, Taylor DP, Rowlands RT, Crawford MS, Lasure LL (1993) Screening microbial metabolites for new drugs - theoretical and practical issues. J Antibiot 46:535–544 6. Höfs R, Walker M, Zeeck A (2000) Hexacyclinic acid, a polyketide from Streptomyces with a novel carbon skeleton. Angew Chem 39:3258–3261 7. Bode HB, Bethe B, Höfs R, Zeeck A (2002) Big effects from small changes: possible ways to explore nature’s chemical diversity. Chembiochem 3:619–627 8. Scherlach K, Hertweck C (2006) Discovery of aspoquinolones A-D, prenylated quinoline-2one alkaloids from Aspergillus nidulans, motivated by genome mining. Org Biomol Chem 4:3517–3520 9. Bills G, Platas G, Fillola A, Jiménez MR, Collado J, Vicente F, Martín J, González A, Bur-Zimmermann J, Tormo JR, Peláez F

10.

11.

12.

13.

14.

15.

(2008) Enhancement of antibiotic and secondary metabolite detection from filamentous fungi by growth on nutritional arrays. J Appl Microbiol 104:1644–1658 Duetz W, Chase M, Bills G (2010) Miniaturization of fermentations. In: Demain A, Davies J, Baltz R (eds) Manual of industrial microbiology and biotechnology, 3rd edn. ASM, Washington, DC, pp 99–116 Duetz WA (2007) Microtiter plates as minibioreactors: miniaturization of fermentation methods. Trends Microbiol 15:469–475 Bills GF, Martín J, Collado J, Platas G, Overy D, Tormo JR, Vicente F, Verkleij G, Crous P (2009) Measuring the distribution and diversity of antibiosis and secondary metabolites in the filamentous fungi. Soc Ind Microbiol News 59:133–147 Roemer T, Xu DM, Singh SB, Parish CA, Harris G, Wang H, Davies JE, Bills GF (2011) Confronting the challenges of natural productbased antifungal discovery. Chem Biol 18: 148–164 Vicente F, Basilio A, Platas G, Collado J, Bills GF, Del Val AG, Martin J, Tormo JR, Harris GH, Zink DL, Justice M, Kahn JN, Pelaez F (2009) Distribution of the antifungal agents sordarins across filamentous fungi. Mycol Res 113:754–770 Bills GF, Platas G, Overy D, Collado J, Fillola A, Jiménez MR, Martín J, del Val AG, Vicente F, Tormo JR, Peláez F, Calati K, Harris G, Parish C, Xu D, Roemer T (2009) Discovery of the parnafungins, antifungal metabolites

78

J.R. Tormo et al.

that inhibit mRNA poly-adenylation, from the Fusarium larvarum complex and other hypocrealean fungi. Mycologia 101:445–469 16. Peláez F, Collado J, Platas G, Overy DP, Martín J, Vicente F, González del Val A, Basilio A, de la Cruz M, Tormo JR, Fillola A, Arenal F, Villareal M, Rubio V, Baral HO, Galán R, Bills GF (2010) The phylogeny and intercontinental distribution of the pneumocandin-producing anamorphic fungus Glarea lozoyensis. Mycology 2:1–17 17. Ondeyka J, Harris G, Zink D, Basilio A, Vicente F, Bills G, Platas G, Collado J, González A, de la Cruz M, Martin J, Kahn JN, Galuska S, Giacobbe R, Abruzzo G, Hickey E, Liberator P, Jiang B, Xu DM, Roemer T, Singh SB (2009) Isolation, structure elucidation, and biological activity of virgineone from Lachnum virgineum using the genome-wide

Candida albicans fitness fest. J Nat Prod 72:136–141 18. Herath K, Harris G, Jayasuriya H, Zink D, Smith S, Vicente F, Bills G, Collado J, Gonzalez A, Jiang B, Kahn JN, Galuska S, Giacobbe R, Abruzzo G, Hickey E, Liberator P, Xu DM, Roemer T, Singh SB (2009) Isolation, structure and biological activity of phomafungin, a cyclic lipodepsipeptide from a widespread tropical Phoma sp. Bioorg Med Chem 17:1361–1369 19. Xu D, Ondeyka J, Harris GH, Zink D, NielsenKahn J, Wang H, Bills G, Platas G, Wang W, Szewczak AA, Liberator P, Roemer T, Singh SB (2011) Isolation, structure and biological activities of fellutamides C and D from an undescribed Metulocladosporiella (Chaetothyriales) using the genome-wide Candida albicans fitness test. J Nat Prod 74(8):1721–1730. doi:10.1021/np2001573

Chapter 4 - Springer Link

Chapter 4 - Springer Link

Suggest Documents